Optimization of Inconel 718 Downskin Parameters for the Production of Metallic Parts Via Laser Powder Bed Fusion

These days, Additive Manufacturing (AM) technologies are considered as advanced processes in which it is possible to produce complex shape components in a layer by layer manner. It is interesting to notice that in these technologies, it is reported that in the production of the parts with angles higher than 45°, no support is required. Whereas, below this angle, the presence of support that can counterbalance the recoating blade’s force and dissipates heat is necessary. In fact, there is a risk of detachments that can cause the piece failure in these angles and increase the heavy dross formation on the downskin surface (high roughness). However, through the optimization of some parameters, it is possible to decrease this angle’s value. Therefore, the thesis topic was to find the optimized downskin parameters for IN718 alloy to improve the overhang surfaces’ quality in inclined specimens. The work started with in-depth bibliographic research on downskin parameters. The most critical parameters were found to be overhang angle, laser power, laser speed, hatch distance, and the number of layers processed with the downskin parameters. Based on the knowledge acquired, the parameters’ optimization was performed in Prima Industrie SpA using a Print Sharp machine. The experimental procedures consisted of three "design of experiments" (DoE) and a repeatability test for the first one. The first DoE was made by a 3^3 factorial experiment for specimens inclined at 30°, 35°, and 40°, modifying the laser power, the laser speed, and the hatch distance. The roughness analysis on the downskin surface was used as a key performance indicator. As a result, the best eight sets of parameters (for angles at 35° and 40°) with a downskin roughness lower than 21 μm were found (literature value at 45° for Inconel 718 is 19 μm). To verify the results' accuracy, a repeatability test was performed by printing and analyzing some specimens using the same parameters. The variability detected was always lower than 5%, confirming the consistency of the results. The second DoE aimed to evaluate the porosity using image analysis in which the specimens were cut, polished, and subsequently analyzed using an optical microscope. The densities of the specimens were always higher than 99.2% for the best sets of parameters. For this reason, it is not expected a change in mechanical characteristics in the downskin region. Finally, the third DoE was performed to evaluate the specimens' tensile properties printed with the best set of downskin parameters (lowest roughness). The core parameters during the printing phase have not been modified, and so it is expected that the mechanical properties of the specimens did not change. All in all, in this thesis, the most influencing downskin parameters and their best combination for IN718 samples are found and reported. The outcomes demonstrated that it is possible for the IN718 to print up to 35° without the need for the supports, keeping high density, and roughness lower than 18 μm. This result allows a decrease in time and costs since are no more needed support structures. Furthermore, are now possible new design optimizations for the IN718, such as new lattice structures, applications with design constraints (rotor with angles lower than 40°), and internal features (where supports removal is impracticable).